DE60027586T2 - METHOD FOR IMPROVING THE COLD STUMBBILITY OF AN ELECTROCHEMICAL FUEL CELL - Google Patents

Abstract

Temperature dependent methods can be used to improve the cold start capability of fuel cell electric power generation systems. A method of ceasing operation of an electric power generation system improves the cold start capability and freeze tolerance of a fuel cell stack by reducing the amount of water remaining within the passages of the stack. The method involves purging one or more of the fuel cell stack oxidant and fuel passages at shutdown prior to allowing the fuel cell stack to drop to temperatures below the freezing point of water. Preferably purging at shutdown is conducted at a temperature below the stack operating temperature. Another method, used at start-up, involves directing a coolant fluid stream to the fuel cell stack only after a predetermined temperature above the freezing temperature of water is exceeded. Preferably, after freezing the fuel cell stack is heated to a temperature above its normal operating temperature before operation is commenced.

Description

Territory of invention

The The present invention relates to methods for improving cold start capability an electrochemical fuel cell. More specifically, the present concerns Invention temperature-dependent Method for improving the cold start capability of fuel cell power generation systems, comprising a fuel cell stack.

background the invention

electrochemical Fuel cells convert fuel and oxidants into electricity and reaction products around. Use solid state polymer electrochemical fuel cells generally a membrane electrode assembly or membrane-electrode assembly ("MEA" - Membrane Electrode Assembly), an ion exchange membrane or a solid polymer electrolyte which is disposed between two electrodes, which typically a layer of a porous, electrically conductive sheet material include carbon fiber paper or carbon cloth. The MEA contains a catalyst layer, typically in the form of finely ground Platinum, at each membrane / electrode interface, to the desired electrochemical Induce reaction. In operation, the electrodes are electrical coupled together to form a circuit for conducting electrons between the electrodes to provide an external circuit.

At At the anode, the fuel flow moves through the porous anode substrate and is oxidized at the anode electrocatalyst layer. At the Cathode, the oxidant flow moves through the porous cathode substrate and is reduced at the cathode electrocatalyst layer to to form a reaction product.

For fuel cells that use hydrogen as the fuel and oxygen-containing air (or substantially pure oxygen) as the oxidant, the catalyzed reaction at the anode generates hydrogen cations (protons) from the fuel supply. The ion exchange membrane facilitates the migration of protons from the anode to the cathode. In addition to conducting protons, the membrane isolates the hydrogen-containing fuel stream from the oxygen-containing oxidant stream. At the cathode electrocatalyst layer, the oxygen reacts with the protons that have passed through the membrane to form water as a reaction product. The anode and cathode reactions in hydrogen / oxygen fuel cells are shown in the following equations:
Anode reaction: H ₂ → 2H ⁺ + 2e ^-
Cathode reaction: 1 / 2O ₂ + 2H ⁺ + 2e ^- → H ₂ O

at typical fuel cells, the MEA is between two electric conductive fluid flow field plates or separating plates arranged. Fluid flow field plates have at least a flow channel formed in at least one of its main planar surfaces is. The flow channels direct the fuel and the oxidizer to the respective ones Electrodes, namely the anode on the fuel side and the cathode on the oxidant side. The fluid flow field plates act as current collectors, provide a holder for the electrodes, provide access channels for the Fuel and the oxidizing agent to the respective anode and cathode surfaces ready and see channels for removing reaction products, such as water, the while the operation of the cell are formed. In the surfaces of Partition plates are typically not formed flow channels, however, they are combined with an adjacent layer of material used the access channels for the Fuel and the oxidizing agent to the respective anode and Cathode electrocatalyst and channels for removing reaction products provides. The preferred operating temperature range for solid polymer fuel cells is typically between 50 ° C and 120 ° C, especially typical between about 75 ° C and 85 ° C.

Two or more fuel cells may be electrically connected in series with each other to increase the overall output power of the assembly. In series arrangements, one side of a given fluid flow field or separation plate may serve as the anode plate for one cell and the other side of the fluid flow field or separation plate as the cathode plate for the adjacent cell. Such a multiple fuel cell assembly is referred to as a fuel cell stack and is usually held together in its assembled state by tie rods and end plates. The stack typically includes inlet ports and manifolds for directing the fluid fuel stream (such as substantially pure hydrogen, methanol reformate or natural gas reformate, or a methanol-containing stream in a direct methanol fuel cell) and the fluid oxidant stream (such as substantially pure oxygen , oxygen-containing air or oxygen in a carrier gas, such as nitrogen) to the individual fuel cell reactant flow channels. The stack also typically includes an inlet port and a manifold for directing a flow of coolant, typically water, to internal passages in the stack to absorb heat generated by the fuel cell during operation. The stack also includes substantially outlet manifolds and outlet ports for expelling the depleted reactant streams and the reaction products, such as water, and an outlet manifold and outlet port for the coolant stream exiting the stack. In a power generation system, various fuel, oxidant, and coolant lines transport these fluid streams toward and away from the fuel cell stack.

If an electrical load (with one or more load elements) in an electrical circuit is placed, the electrodes connecting the fuel and the oxidizer directly proportional to the electric power consumed by the load is subtracted, causing the ohmic resistance of the load changed becomes.

Solid polymer fuel cells generally use perfluorosulfone ion exchange membranes, such as those of DuPont under the trade name NAFION and marketed by Dow under the trade designation XUS 13204.10. at Using such membranes, the fuel and oxidant reactant streams typically become Moistened before they are incorporated in solid polymer fuel cells to to facilitate proton transport through the ion exchange membrane and dehydration (and damage) to avoid the membrane separating the anode and cathode of each cell.

Everyone emerging from the fuel cell stack reactant stream contains generally water. The fuel stream leaving the anodes contains the water used to humidify the incoming fuel stream was added, as well as any product water coming from the cathode was pulled out through the membrane. The emerging from the cathodes Contains oxidant stream generally the water used to wet the incoming oxidant stream was added, as well as formed at the cathode product water.

at many fuel cell applications, such as drive applications, It may be necessary or desired be the operation of a solid polymer electrolyte fuel cell stack to start when the stack core temperature is below the freezing temperature of water lies. As used herein with the freezing temperature of water the freezing temperature of free water meant, that is 0 ° C at 1 atmosphere. At the Stopping the operation of the solid polymer fuel cell stack it as well necessary or desired be, the cold startability and freeze tolerance of the pile through Reduce the in the fuel, oxidant and coolant channels of the Stack of remaining water to improve. When freezing expands in the stack channels remaining water can and structures within the stack, such as e.g. the membrane / electrocatalyst interface, the reactant channels, conduits and seals as well as the porous ones Electrode substrate material, potentially damaging.

If It is expected that a solid polymer fuel cell stack low temperatures will be exposed, especially temperatures below the freezing temperature of water, one or more special Startup and shutdown techniques are used. These techniques can be the Cold startability and freeze tolerance of the stack and the later fuel cell performance improve. A measure of Performance of an electrochemical fuel cell is in the Voltage output of the cell at a given current density. A higher Performance comes with a higher one Voltage output at a given current density or higher current density associated with a given voltage output.

Summary the invention

One Method for starting the operation of a power generation system accelerates the heating of the fuel cell stack to the desired operating temperature range. The system includes a fuel cell stack that is connected to an external fuel cell electrical circuit is connectable to the external circuit to supply with electric current. The stack comprises at least a fuel cell, wherein the at least one fuel cell a membrane electrode assembly having an anode, a cathode and an ion exchange membrane disposed between the anode and the cathode having. The system further includes a fuel stream and a Oxidant stream, wherein each of the streams may flow to the fuel cell stack. The system further includes a coolant flow in thermal contact can flow with the fuel cell stack.

The Rated operating temperature of the stack can be directly (e.g., by placing a temperature sensor at one or more locations in the stack) or be measured indirectly, e.g. by monitoring the temperature of a or more of the fluid streams leaving the stack. In the Practice can such measurements are used to represent a representative or approximate To provide or derive the value of the stack operating temperature.

The temperature threshold at which the coolant flow is started is preferably more than about 0 ° C, but may also be below the typical desired operating temperature range of the fuel cell stack. For example, the threshold could be between about 30 ° C and 50 ° C or are within the desired operating temperature range, which is typically between about 75 ° C and 85 ° C for a solid polymer fuel cell. Once the desired operating temperature range has been reached, conventional temperature control techniques may be used thereafter to further operate the fuel cell stack within the desired temperature range.

This Method is particularly suitable for starting the operation when at least a portion of the membrane electrode assembly is at a temperature below the freezing temperature of water has.

The predefined temperature threshold at which the coolant flow is started, is higher than the normal desired Operating temperature of the stack. It is preferably, for example at least about 10 ° C above the normal desired operating temperature of the pile. For a typical solid polymer fuel cell For example, the preferred operating temperature range may be approximately between 75 ° C and 85 ° C lie. In this embodiment of the procedure could the coolant flow delayed until the operating temperature reaches a value between about 95 ° C and 105 ° C.

It has been shown to be in situ "heat treatment" of a membrane electrode assembly after a cold start in certain situations, the later fuel cell performance a fuel cell opposite improved starting of the operation, in which the cell is not above her normal operating temperature range is operated. This improved Method is again particularly suitable for starting the operation, when at least a portion of the membrane electrode assembly is at a temperature below the freezing temperature of water and in particular at a Operation with air (instead of a substantially pure oxidant). Operationally, the in-situ heat treatment process, next to the delay the coolant flow, can be achieved in many other ways.

Therefore comprises a second method for starting the operation of a power generation system a period of time while the stack over his normal operating temperature is operated. The system includes one Fuel cell stack, with an external electrical circuit connectable to the external circuit with electric current to supply. The stack comprises at least one fuel cell, wherein the at least one fuel cell is a membrane electrode assembly with an anode, a cathode and one between the anode and the cathode has arranged ion exchange membrane. The system further comprises a fuel stream and an oxidant stream, where each of the streams flow to the fuel cell stack can. The system optionally further comprises a coolant stream, in thermal contact can flow with the fuel cell stack.

The method comprises:
Supplying the external circuit with electric current from the fuel cell stack so that the temperature of the at least one fuel cell rises to a temperature that is above the normal operating temperature range of the stack, and
Reduce the operating temperature of the stack to a temperature within the normal operating temperature range.

The over the normal operating temperature lying temperature is typically predefined. In preferred embodiments of the method At times, the fuel cell stack will be at least about 10 ° C above its normal desired Operating temperature operated. In a typical solid polymer fuel cell the preferred operating temperature range e.g. between about 75 ° C and 85 ° C, so that the stack is preferably over operated at a value between about 95 ° C and 105 ° C for a certain period of time can be resumed before operating within the range between about 75 ° C and 85 ° C becomes. The time span while the stack with the higher one Temperature is operated, may be variable or it may be to trade for a predefined period of time. The stack can, for example for about 1-2 minutes or a few seconds at the higher temperature become. The temperature and / or the period of the operating phase with higher Temperatures can however in dependence supervised by some Operating parameters of the fuel cell system can be adjusted. The preferred temperature and / or time span may be, for example, from the Temperature of the surrounding environment, the humidity conditions within the stack, a reactant quality or purity indicating how long the stack is at a low tempera ture has been stored, or a the fuel cell performance indicating electrical Parameter dependent be.

As an alternative to the above heat treatment process, there may be some advantages over heating the fuel cell from below the freezing point of water to above its normal operating temperature range prior to starting the operation thereof. For example, an externally driven heater could be used to heat the stack or coolant circulating through the stack, or a hot fluid stream could be used from elsewhere in the system. Preferably For example, during the heat treatment, a gas stream is circulated through one or both of the reactant flow channels.

at the above method leads The exothermic operation of the stack often causes the operating temperature of the pile rises. It can however about it In addition, other means are used to increase the temperature to within or above the desired stack operating temperature range to accelerate or facilitate.

Summary the drawings

1 Figure 4 is a side exploded view of a typical solid state polymer electrochemical fuel cell having a membrane electrode assembly disposed between two fluid flow field plates.

2 FIG. 4 is a spatial sectional view of an electrochemical fuel cell stack. FIG.

3 is a schematic representation of a fuel cell power generation system.

4 FIG. 3 is a mixed plot of fuel cell stack voltage vs. time in minutes (Graph A) and fuel cell stack core temperature versus time in minutes (Graph B) of a 10-cell stack that has been started after the stack has reached a core temperature of. FIG. 11 ° C had leveled.

5 FIG. 12 is a mixed plot of fuel cell stack voltage versus time in minutes (Graph C) versus fuel cell stack core temperature versus time in minutes (Graph D) of a 4-cell stack started operating at a core temperature of -19 ° C.

6 Figure 3 is a graph of voltage versus current density of a fuel cell containing a membrane electrode assembly with a Dow-Pont ^™ membrane and subjected to three cold rinse freeze / thaw cycles.

7 is a graph of voltage as a function of current density of a fuel cell containing a membrane electrode assembly with a Nafion ^® -1135 membrane and three cold purge freeze / thaw cycles was subjected.

8th is a graph of voltage as a function of current density of a fuel cell containing a membrane electrode assembly with a Nafion ^® -1135 membrane and a cold purge freeze / thaw cycle shorter duration has been subjected.

9 Figure 3 is a graph of voltage vs. current density of a fuel cell containing a membrane electrode assembly with a Dow-Pont ^™ membrane and subjected to three warm-rinse freeze / thaw cycles.

10 is a graph of voltage as a function of current density of the fuel cell containing a membrane electrode assembly with a Nafion ^® -1135 membrane and three hot purge freeze / thaw cycles was subjected.

11 is a graph of cell voltage of the fuel cell containing a membrane electrode assembly with a Nafion ^® -1135 membrane, after each of a series of 55 cold purge freeze / thaw cycles.

12 FIG. 12 is a plot of stack voltage versus operating time of an 8-cell Ballard-Mark 513 fuel cell stack that has undergone a series of freeze-thaw cycles incorporating four heat treatment cycles.

13 FIG. 12 is a plot of the stack voltage versus current density for generating the data according to FIG 12 used fuel cell stack before and after the heat treatment.

Precise description of the preferred embodiments

1 shows a typical fuel cell 10 , The fuel cell 10 includes a membrane electrode assembly 12 between an anode flow field plate 14 and a cathode flow field plate 16 is arranged. The membrane electrode assembly 12 consists of an ion exchange membrane 20 between two electrodes, namely the anode 21 and the cathode 22 , is arranged. In conventional fuel cells, the anode comprises 21 and the cathode 22 each a substrate of a porous, electrically conductive layer material 23 respectively. 24 , eg carbon fiber paper or carbon fabric. On each substrate there is a thin layer of electrocatalyst on a surface thereof 25 respectively. 26 at the interface with the membrane 20 arranged to make each electrode electrochemically active.

As further in 1 is at least one fuel flow channel 14a in the anode 21 facing surface of the anode flow field plate 14 Engraved, milled or shaped. Similarly, at least one oxidant flow channel is 16a in the cathode 22 facing surface of the cathode separator plate 16 Engraved, milled or shaped. If you have the interacting surfaces of the electrodes 21 and 22 are mounted, form the channels 14a and 16a the reactant flow field channels for the fuel or the oxidant. The flow field plates are electrically conductive.

Referring now to 2 includes a fuel cell stack 100 a plurality of fuel cell assemblies, a number of them in 2 With 111 is designated. Each of the fuel cell assemblies includes a membrane electrode assembly 112 that is between a pair of fluid flow field plates 114 . 116 is arranged. The fuel cell stack 100 also includes a first end plate 130 and a second end plate 140 ,

The plate 130 includes fluid inlet openings 132 . 134 . 136 for introducing fluid fuel, oxidant or coolant streams into the stack. The plate 140 includes fluid outlet openings 142 . 144 . 146 for discharging fluid fuel, oxidant and coolant streams from the stack. The fluid outlet ports are in fluid communication with the corresponding fluid inlet ports via channels provided in the stack.

In the fuel cell assemblies, a series of openings are formed which cooperate with corresponding openings in adjacent assemblies to form fluid manifolds 152 . 154 . 156 . 162 . 164 . 166 in the pile 100 to build. The fluid manifolds are each surrounded by a sealant or seal. In addition, a peripheral seal on the outer circumference of each fuel cell fluid-tightly isolates the inner, electrochemically active portion of the fuel cell from the external environment.

A via the fuel inlet 132 fuel flow entering the stack is via the manifold 152 directed to the individual fuel flow field plates. After passing through the fuel flow field plate channels, the fuel flow in the manifold 162 collected and over the Brennstoffauslassöffnung 142 released from the pile. Similarly, via the oxidant inlet port 134 into the stack entering oxidant stream through the manifold 154 directed to the individual oxidant flow field plates. After passing through the oxidant flow field plate channels, the oxidant stream becomes in the manifold 164 collected and via the Oxidationsmittelauslassöffnung 144 released from the pile. A via the coolant inlet opening 136 introduced, fluid coolant (typically water) is through the manifold 156 to the coolant plate assemblies (not shown) in the stack 100 directed. The coolant flow is in the manifold 166 collected and via the coolant outlet 146 released from the pile. The coolant distributors 156 . 166 can be provided with compliant means (not shown), such as closed-cell foam tube cushions or inserts, to accommodate the expansion of freezing water. drawbars 170 extend between the end plates 130 and 140 to the pile 100 by means of each attached to opposite ends of each pull rod fastening nuts 172 and between the fastening nuts 172 and the endplates 130 . 140 arranged disc springs 174 in its assembled state, compress and secure.

3 is a schematic representation of a fuel cell power generation system 200 with a fuel cell stack 210 , The fuel cell stack 210 includes negative and positive busbar plates 212 respectively. 214 to which an external circuit with a variable load 216 by closing the switch 218 is electrically connected. The system includes a fuel cycle (hydrogen cycle), an oxidant cycle (air cycle), and a coolant water loop. The reactant and coolant flows are in the system in different, schematically in 3 circulated lines shown.

A source of hydrogen 220 is with the stack 210 connected, the pressure by a pressure regulator 221 is controlled. In the one out of the pile 210 Exiting hydrogen flow of existing water is in an impact drum 222 collected by opening the valve 223 can be emptied. Unreacted hydrogen is pumped through 224 in a return loop 225 to the pile 210 recycled. An air source 230 is with the stack 210 connected, with their pressure through the pressure regulator 231 is controlled. In the one out of the pile 210 Exiting airflow existing water is in a container 232 collected, which by opening the valve 233 can be drained, with the air flow through the valve 234 is released from the system.

In the coolant water circuit 240 gets water from the container 232 pumped and by means of the pump 241 through the pile 210 circulated. The temperature of the water is in a heat exchanger 242 set.

A flushing system 250 is used to purge the hydrogen and oxidant channels in the fuel cell stack 210 used with a non-reactive low-humidity gas. The flow of gas (dry nitrogen) from a purge gas source 260 to the hydrogen and air intake lines 261 . 262 is through the valves 263 . 264 and the three-way valves 266 . 267 controlled. The nitrogen pressure is controlled by a pressure regulator 265 controlled.

4 FIG. 12 is a mixed plot of fuel cell stack voltage vs. time in minutes (Graph A) and fuel cell stack core temperature versus time in minutes (Graph B) of a 10-cell stack to which the fuel and oxidant flow was restored after the fuel cell stack core temperature. FIG Stack had settled to a core temperature of -11 ° C.

Of the Stack was previously in operation, so the reactant flow channels were wet Contained gases. Before lowering the stack core temperature below the Freezing temperature of water, the reactant and coolant water channels were in the stack by circulating dry, compressed air therethrough cleaned. The stack core temperature then went below the freezing temperature lowered by water, adding the stack of an environment with a temperature was exposed below the freezing temperature of water was. For the purpose of the examples described herein was typically the stack placed in an insulated chamber, the fluid and power connections of the Stack through the chamber walls were installed. Cold gaseous Nitrogen from a liquid nitrogen source was circulated through the chamber. The stack core temperature was below Using a thermocouple measured in a thermal conductive plate was arranged, extending between two fuel cells in the center of the pile. The stack voltage, the stack current and the ambient temperature were also monitored.

When the circulation of hydrogen and air through the stack at one Stack core temperature of -11 ° C (at the time = 0 minutes), the quiescent voltage was normal. A burden (360 A) was about connected to the circuit for three minutes, which caused that the stack core temperature rose rapidly as the voltage decreased, but gradually recovered. As soon as the operation of the stack was started, the exothermic reaction of hydrogen and oxygen in the stack and resistance heating due to internal ohmic losses Increase in stack core temperature.

5 FIG. 12 is a mixed plot of fuel cell stack voltage versus time in minutes (Graph C) versus fuel cell stack core temperature versus time in minutes (Graph D) of a 4-cell stack started operating at a core temperature of -19 ° C.

In turn were because the stack had previously been in operation before lowered the stack core temperature to -19 ° C was the reaction partner channels Purified in the stack by circulating dry nitrogen. Coolant Water stayed behind in the coolant channels. The Cleaning fluid is preferably a noble gas such as nitrogen. The circulation of hydrogen and air was at connection of a Last (50 A) started. Approximately 2 minutes passed before the output current reached 50A. The Load was increased to 260 A, once the stack is about Reached 30 ° C had, in which case the coolant pump has been activated. A cell in the stack was not working properly and was lying thus below the normal average cell voltages.

During the Starts the batch operation, it has proven to be advantageous to a circulation of the fluid coolant flow in the stack refrain until the stack is over has reached the freezing temperature of water lying temperature. It is particularly preferred not to circulate the fluid coolant stream, until the stack has reached a temperature that is the desired Stack operating temperature is equal to or comes close. In this regard absorbed and transport the circulating fluid coolant stream, provided he was not heated in advance, Heat off, otherwise to warm up the Pile available would. Abandonment of the circulation of the fluid coolant flow accelerates therefore the warming pile on its desired Operating temperatur.

The Cold startability and freezing tolerance of fuel cells can be improved be reduced by reducing the amount of water in the channels of the Stack when setting the operation and lowering the stack core temperature remains close to or below the freezing temperature of water. As used herein, "freeze tolerance" refers to the ability of a fuel cell or a fuel cell stack, after one or more freeze / thaw cycles to maintain substantially the same performance.

The reactant channels, including the manifolds and the individual fuel cell reactant flow channels in a fuel cell stack, are preferably purged with a fluid stream before the temperature of the stack is lowered below the freezing temperature of water. It is preferable to use a fluid that is unreactive in the fuel cell environment, such as gaseous nitrogen. A liquid can be used as flushing fluid. Preferred would be a liquid which does not freeze at the temperature to which the fuel cell is to be exposed and which has no deleterious effect on the fuel cell components. Alternatively, the reactant streams themselves can be used as purge streams. The flushing fluid is preferably, if it is a gas, dry or at least not moistened. Therefore, when the reactant streams are used as purge streams, reactants should be used current humidifier (if any) are bypassed to provide currents whose water carrying capacity is greater than that of humidified reactant streams. The greater water carrying capacity of unhumidified reactant purge streams results in more efficient absorption and removal of water from the reactant stream conduits and porous components of the stack. Although it may be desirable in certain situations to purge all reactant and coolant channels, it has proven effective in many cases to purge only the oxidant flow channels. This can simplify the system and the shutdown sequence.

It has been shown to have improved cold startability and freeze tolerance of fuel cells at several freeze / thaw cycles are also achieved can, if one or more of the fuel, oxidant, coolant and humidification channels rinsed after the stack core temperature returns to normal room temperature or below (hereinafter referred to as "cold rinse"). The beneficial effect the conditioner is not quite so pronounced if the stack channels rinsed at a temperature which are within the normal batch operating temperature range is (hereinafter referred to as "warm rinse").

Examples - rinsing procedure

experimental details

The effect of cold and hot flushes membrane electrode assemblies with two different types of membranes, a Nafion ^® -112- and a DowPont ^™ membrane, in a Ballard Mark 513 fuel cell-single with an internal humidifier was investigated. Separate water supply lines for the coolant and humidification streams were used. The coolant outlet temperature was 85 ° C with a ΔT (temperature change between inlet and outlet) of 10 ° C at 1000 ASF (10764 ASM) using air as the oxidant. Both membrane electrode assemblies (MEAs) had a screen printing anode containing 3.87 mg / cm ² platinum black electrocatalyst on carbon fiber paper. At the cathode, both MEAs had 3.87 mg / cm ² platinum black electrocatalyst, which was applied by hand to carbon fiber paper. The Nafion ^® -1135 membrane used in the MEA no. 513-15 had a relative equivalent weight of 1100 and a thickness of about 85 microns (dry). The DowPont ^™ membrane used in MEA No. 513-22 had a relative equivalent weight of 800 and a thickness of approximately 100 μm (wet).

The Mark 513 cell was assembled and stored overnight at 600 ASF (6458 ASM) each at an air / fuel pressure of 30/30 psig (207/207 kPa Measuring pressure) and a stoichiometry operated by 2 / 1.5. The fuel was essentially pure hydrogen. "Stoichiometry" is the ratio of the Fuel cell stack supplied Reaction partner quantity to actually consumed in the fuel cell stack reaction partner. In In this case, a fuel stoichiometry of 1.5 means that the Fuel cell for in each case 100 parts actually hydrogen consumed in the fuel cell contains 150 parts of hydrogen supplied become.

Cold purge freeze / thaw cycles

In the initial series of three freeze / thaw cycles (results are in the 6 and 7 shown), the cell was cooled from its normal operating temperature (about 85 ° C) to room temperature (about 23 ° C) prior to rinsing. In each case, the fuel, oxidizer, coolant and humidification passages were purged with nitrogen for approximately 7 minutes. The Nafion ^® -1135 membrane-containing cell was a fourth freeze / thaw cycle with a purge duration of only approximately 1 minute subjected (results are in 8th shown). The cell inlets and outlets were covered and the cell placed in a freezer. The internal sealing pressure within the cell was maintained during the freezing process. The temperature of the freezer was about -20 ° C. The duration of the freezing process was between 15 and 20 hours. After removal from the freezer, the coolant lines were connected and the cell was heated to 50 ° C. At this point, fuel cell operation was started at 50 ASF (538.2 ASM) at inflated fuel and oxidant flow rates. When the cell temperature reached 60 ° C, the current density was increased to 600 ASF (6458 ASM) and the cell operated for at least one hour or until the cell voltage stabilized. A polarization test was performed between 0 and 1000 ASF (0 and 10764 ASM) on each of the two MEAs tested using two different oxidant streams: air and substantially pure oxygen.

) und jeweils nach den vorstehend beschriebenen drei Kaltspülungs-Gefrier-/Tauzyklen. Im Hinblick auf jeden der Oxidationsmittelströme sind die vier Diagramme in 6 nur schwer voneinander zu unterscheiden. 6 FIG. 12 is a graph of voltage vs. current density of the MEA 513-22 (DowPont ^™ membrane) containing fuel cell. FIG. Diagrams 1-4 show the performance of using air before freezing (represented by the solid line and full data points graphs ♦) and after each of the three cold rinse freeze / thaw cycles described above. Diagrams 5-8 show the performance of using oxygen prior to freezing (represented by the diagram with FIG drawn line and the full data points

) and after each of the three cold rinse freeze / thaw cycles described above. With regard to each of the oxidant streams, the four diagrams are in 6 difficult to distinguish from each other.

) und jeweils nach den vorstehend beschriebenen drei Kaltspülungs-Gefrier-/Tauzyklen. Wiederum sind im Hinblick auf jeden der Oxidationsmittelströme die vier Diagramme in 7 nur schwer voneinander zu unterscheiden. 7 is a plot of voltage as a function of the current density of an MEA 513-12 (Nafion ^® -1135 membrane) containing fuel cell. Diagrams 1-4 show the performance of using air before freezing (represented by the solid line and full data points graphs ♦) and after each of the three cold rinse freeze / thaw cycles described above. Diagrams 5-8 show the performance when using oxygen before freezing (represented by the solid line graph and the full data points

) and after each of the three cold rinse freeze / thaw cycles described above. Again, with respect to each of the oxidant streams, the four diagrams in FIG 7 difficult to distinguish from each other.

) und nach dem vorstehend beschriebenen vierten Kaltspülungs-Gefrier-/Tauzyklus. Wiederum sind im Hinblick auf jeden der Oxidationsmittelströme die zwei Diagramme in 8 nur schwer voneinander zu unterscheiden, was darauf hinweist, dass eine kürzere Spüldauer zufriedenstellende Ergebnisse zu erzielen vermag. 8th is a plot of voltage as a function of the current density of an MEA 513-12 (Nafion ^® -1135 membrane) containing fuel cell. Diagrams 1-2 show the performance of using air before freezing (represented by the solid line and full data point graphs ♦) and after the fourth cold rinse freeze / thaw cycle described above, in which a shorter cleaning time (approx 1 minute) was used. Diagrams 3-4 show the performance of using oxygen before freezing (represented by the solid line graph and the full data points

) and after the fourth cold rinse freeze / thaw cycle described above. Again, with regard to each of the oxidant streams, the two diagrams in FIG 8th difficult to distinguish from one another, indicating that a shorter purging time can produce satisfactory results.

Thus, based on the in the 6 . 7 and 8th showed substantially no mass transfer losses during the series of three or four freeze / thaw cycles for both MEAs. Performance was maintained at approximately (before freezing) polarization levels after each freeze / thaw cycle. Both MEAs thus showed a favorable freezing / dewing tolerance when using the cold rinsing technique.

Hot purge freeze / thaw cycles

at a later one A series of three freeze / thaw cycles was performed on each cell before cooling Stack operating temperature (about 85 ° C). The Fuel, oxidant, coolant and humidification channels approximately Purged with nitrogen for 1 minute. The cell inlets and outlets became covered and the cell placed in a freezer. The internal sealing pressure inside the cell was during maintained the freezing process. The temperature of the freezer was about -20 ° C. The duration the freezing time was between 15 and 20 hours. After Removal from the freezer were the coolant lines connected, the cell heated to operating temperature and the operation started using the substantially same process, used in the cold rinse freeze / thaw cycles described above has been. A polarization test was between 0 and 1000 ASF (0 and 10764 ASM) at each of the two tested MEAs again using two different Oxidant streams carried out: Air and essentially pure oxygen.

) und jeweils nach den vorstehend beschriebenen drei Warmspülungs-Gefrier-/Tauzyklen. Bei höheren Stromdichten scheint bei Verwendung von Luft nach dem dritten Gefrierzyklus ein signifikanter Stofftransporteffekt aufzutreten, zu urteilen nach dem zunehmenden Unterschied zwischen den Luft- und Sauerstoffleistungsniveaus. 9 FIG. 12 is a graph of voltage vs. current density of the MEA 513-22 (DowPont ^™ membrane) containing fuel cell. FIG. Diagrams 1-4 show the performance of using pre-freezing air (represented by the solid line and full data point graphs ♦) and each after the three hot-rinse freeze-thaw cycles described above. Diagrams 5-8 show the performance when using oxygen before freezing (represented by the solid line graph and the full data points

) and each after the above-described three warm rinse freeze / thaw cycles. At higher current densities, using air after the third freezing cycle appears to have a significant mass transport effect, judged by the increasing difference between the air and oxygen power levels.

) und jeweils nach den vorstehend beschriebenen drei Warmspülungs-Gefrier-/Tauzyklen. Wiederum scheint bei höheren Stromdichten bei Verwendung von Luft ein signifikanter Stofftransporteffekt aufzutreten, zu urteilen nach dem progressiv zunehmenden Unterschied zwischen den Luft- und Sauerstoffleistungsniveaus nach jedem Gefrier-/Tauzyklus basiert. 10 is a plot of voltage as a function of the current density of an MEA 513-12 (Nafion ^® -1135 membrane) containing fuel cell. Diagrams 1-4 show the performance of using pre-freezing air (represented by the solid line and full data point graphs ♦) and each after the three hot-rinse freeze-thaw cycles described above. Diagrams 5-8 show the performance when using oxygen before freezing (represented by the solid line graph and the full data points

) and each after the above-described three warm rinse freeze / thaw cycles. Again, at higher current densities using air, there appears to be a significant mass transport effect, judged by the progressively increasing difference between the air and oxygen performance levels after each freeze / thaw cycle.

The through the cold rinse technique achieved, particularly advantageous results were further by the following exam support in which a single fuel cell undergoes 55 freeze / thaw cycles was, using the rinse technique only with the coolant channels and cathode-side channels was applied.

experimental details

The effect of repeated cold rinses to a membrane electrode assembly with a Nafion ^® -112 membrane in a Ballard Mark 513 fuel cell-single with an external humidifier was investigated. Separate water supply lines for the coolant and humidification streams were used. The coolant inlet temperature was 70 ° C with a ΔT (temperature change between inlet and outlet) of 15 ° C at 1 A / cm ² using air as the oxidant. The MEA had a screen-printed anode containing 0.34-0.38 mg / cm ² platinum black electrocatalyst and a screen-printed cathode containing 0.73-0.82 mg / cm ² platinum black electrocatalyst on carbon fiber paper, both with Nafion spray coating (0.2 mg / cm ² ).

The Cell was in a temperature controlled environmental chamber respectively at an air / fuel pressure of 27 / 27psig (186/186 kPa gauge pressure) and a stoichiometry operated by 1.8 / 1.2. The fuel was a simulated methanol reformate stream (Composition: 63.5% hydrogen, 22.5% carbon dioxide, 13% nitrogen, 1% methanol and 40 ppm carbon monoxide), where at the anode a vent of 4% was applied. The fuel and oxidant streams were moistened.

For the series of 55 freeze / thaw cycles (results are in 11 shown), the cell was cooled from its normal operating temperature (about 80 ° C) to a chamber temperature at which no part of the stack was below 0 ° C, but the cell temperature before rinsing was about 30 ° C. In each case, the oxidizer channels were cleaned for about 10 seconds with dry (non-humidified) air. The cell inlets and outlets were closed by actuated valves and the temperature in the chamber lowered to about -25 ° C. The duration of each freezing process was about 1 hour. The internal sealing pressure within the cell was maintained during the freezing process. The cell was then thawed to 5 ° C and then heated to 65 ° C by circulating a warm coolant. At this point, operation of the fuel cell was started at 0.5 A / cm ² for 60 minutes, then at 1.0 A / cm ^{2 for} 30 minutes and then a second time at 0.5 A / cm ^{2 for} 30 minutes ,

11 shows the results obtained after each of the 55 such cycles, the cell voltage being measured once it has stabilized to 1.0 A / cm ² during the 30 minutes of operation at this current density (Chart A) and once it is at 0, 5 A / cm ² during the second period of operation had stabilized at this current density (diagram B). At both current densities, the performance degradation over the 55 cycles was negligible: about -0.1 mV / cycle at 0.5 A / cm ² and about -0.2 mV / cycle at 1.0 A / cm ² .

Examples - Heat treatment process

12 FIG. 12 shows a plot of stack voltage vs. operating time of an 8 cell Ballard Mark 513 fuel cell stack that has undergone a series of freeze / thaw cycles of operation. Prior to freezing, the fuel, oxidant and coolant channels were purged with dry gas. During the freezing cycles, the cell inlets and outlets were covered and the cell placed in a freezer. The internal sealing pressure within the cell was maintained during the freezing process. The temperature of the freezer was about -20 ° C. The duration of the freezing process was more than 12 hours in each case. After a few cycles, the stack was operated normally and after further cycles, the stack operating temperature was raised to a temperature above its normal operating temperature for a period of time before normal operation resumed. The stack was operated using humidified air and hydrogen in both cases at 30 psig (207 kPa gauge pressure), at stoichiometries of 2.0 and 1.5, respectively, at a current density of 700 ASF (7535 ASM) in the 12 generate generated data. The coolant inlet temperature was 75 ° C at a ΔT (temperature change between inlet and outlet) of 10 ° C at 1000 ASF (10764 ASM).

Referring to 12 The stack was operated between 425 and 882 hours, frozen several times and then started after heating to 50 ° C. The data points obtained directly after a freeze cycle are marked F. It can be seen that the cell performance has deteriorated after each freezing cycle. After 882 hours, the stack was started after several freeze cycles and heated to only 10 ° C. After 950 hours, the stack was started after several freeze cycles and heated to just 0 ° C. The observed power losses did not appear significantly affected by the starting temperature to have been flown.

A significant improvement in performance after freezing was achieved in 4 cases, in which batch mode started and the stack operating temperature then one over the normal stack operating temperature of about 85 ° C has been increased, namely about 100 ° C. The directly achieved after such heat treatments Data points are marked H.

13 FIG. 12 shows a plot of the stack voltage as a function of the current density of the 8 cell Ballard Mark 513 fuel cell stack used to generate the data according to FIG 12 has been used. Diagrams A, B and C show the power curves when operating with air and the diagrams D, E and F when operating with oxygen. Diagrams A and D show the stacking performance prior to one of the heat treatments referred to in the description of FIG 12 However, after the final freezing cycle, diagrams B and E show the stacking power immediately after the fourth heat treatment and the charts C and F the stacking power about 2 days after the fourth heat treatment. The results using air show a significant and lasting improvement in post-freezing performance after heat treatment. It is believed that the improvement is attributable to the improved mass transfer in the MEA, which is based on the fact that the performance was not significantly affected by the use of oxygen by the heat treatment. This indicates that post freezing performance loss is due, at least in part, to the mass transport problems in the cells, which have a greater effect on air than on a substantially pure oxidant stream. These effects could be due to residual water in the membrane electrode assembly. It is possible that the heat treatment process aids in the removal of residual water from the membrane electrode assembly and thereby improves startup performance, particularly when using air.

Claims (9)

Method for starting the operation of a power generation system with a fuel cell stack connected to an external electrical circuit connectable to the external circuit with electric current to supply, wherein the stack comprises at least one fuel cell and the at least one fuel cell is a membrane electrode assembly with an anode, a cathode and one between the anode and the cathode has disposed ion exchange membrane, wherein the System further includes a fuel stream and an oxidant stream includes and each of the streams flow to the fuel cell stack can, wherein the method comprises: Supplying the external circuit with electric current from the fuel cell stack, so that the temperature the at least one fuel cell rises to a temperature, the above the normal operating temperature range of the stack, and Reduce the operating temperature of the stack to a temperature within the normal operating temperature range. The method of claim 1, wherein the above normal operating temperature range of the stack lying temperature at least 10 ° C above normal Operating temperature range is. The method of claim 1, wherein the above normal operating temperature range of the stack lying temperature between 95 ° C and 105 ° C. The method of claim 1, wherein the above normal operating temperature range of the stack lying temperature in accordance with at least one supervised Operating parameters of the system is set. The method of claim 1, wherein the stack via a predefined period of time above the normal operating temperature range the stack is operated temperature. The method of claim 5, wherein the stack via a Time span of about 1 to 2 minutes with the over the normal operating temperature range of the stack temperature is operated. Method according to Claim 1, in which the period of time in the stack with the over the normal operating temperature range of the stack temperature is operated, in accordance with at least one supervised Operating parameters of the system is set. Method according to claim 1, where the system further comprising a coolant stream, in thermal contact can flow with the fuel cell stack, the method includes: Starting the flow of the coolant flow in thermal contact with the fuel cell stack only after the operating temperature of the stack exceeds the normal operating temperature range of the stack. The method of claim 8, wherein the flow of coolant is continued at least until the operating temperature of the stack in the nor paint operating temperature range has decreased.